VOL. 45, 1959 BIOCHEMISTRY: MARKERT AND M0LLER 753 University of Virginia. This research was supported in part by grants from the National Institutes of Health and the National Science Foundation. I Cleveland, L. R., Mem. Am. Acad. Arts and S&i., 17, 187 (1934). 2 Cleveland, L. R., J. Morph., 85, 197 (1949); 86, 185 (1950); 86, 215 (1950); 87, 317 (1950); 87, 349 (1950); 88, 199 (1951); 88, 385 (1951); 91, 269 (1952); 93, 371 (1953); 95, 189 (1954); 95, 213 (1954); 95, 557 (1954); 97, 511 (1955); Arch. f. Protistenk., 101, 99 (1956). 3 Cleveland, L. R., J. Protozool., 4, 168 (1957). 4Butenandt, A., and P. Karlson, Z. Naturforsch., 9b, 389 (1954). 5 Nutting, W. L., and L. R. Cleveland, J. Exp. Zool., 137, 13 (1958). 6 Cleveland, L. R., and W. L. Nutting, J. Exp. Zool., 130, 485 (1955). 7Cleveland, L. R., Biol. Bull., 48, 455 (1925). MULTIPLE FORMS OF ENZYMES: TISSUE, ONTOGENETIC, AND SPECIES SPECIFIC PA TTERNS* BY CLEMENT L. MARKERT AND FREDDY MOLLER DEPARTMENT OF BIOLOGY, THE JOHNS HOPKINS UNIVERSITY, BALTIMORE Communicated by B. H. Willier, March 30, 1959 The fundamental metabolic activities of organisms are very similar and conse- quently enzymes catalyzing identical reactions may be found in many different organisms and in many different tissues within an organism. When subjected to a variety of physical,'-4 chemical,5' 6, 7 or serological8-'7 tests enzymes from different organisms are commonly found to be different from each other even though catalyzing the same chemical reaction. In view of the demonstrated genetic control of protein synthesis18 it is not surprising that differences should exist in the structure of homologous enzymes or proteins synthesized by animals of different species'9-24 or even by animals of different genotype within the same species.25-29 Rather surprising, however, is the evidence demonstrating that several enzymes exist in multiple molecular forms not only within a single organism but even within a single tissue. Among the enzymes that have been reported to exist in separate molecular types within a single tissue are esterase,30, 1 ribo- nuclease,32' 33' 34 pepsin,35 chymotrypsin,36 trypsin, 7 lysozyme,38 39 cytochrome C,40 xanthine dehydrogenase,4' malate dehydrogenase,424'444 and lactate dehy- drogenase.2-4, 42-49 Likewise, in yeast distinct molecular types of phospho- glyceraldehyde dehydrogenase5W and of enolase5' have been identified. The exist- ence of each of these enzymes as a family of closely related but distinguishable molecular types suggests the need for an extension of the classification of enzymes beyond that based on substrate specificity alone. We propose, therefore, to use the term isozyme to describe the different molecular forms in which proteins may exist with the same enzymatic specificity. In the present investigation three dehydrogenase enzymes-lactate dehydro- genase (LDH),t malate dehydrogenase (MDH), and isocitrate dehydrogenase (IDH)-have been resolved into physically distinct forms, that is, into isozymes. In addition to the dehydrogenases, three enzymes with broad substrate specifici- Downloaded by guest on September 27, 2021 754 BIOCHEMISTRY: MARKERT AND MOLLER PROC. N. A. S. ties-that is, alkaline phosphatase, peroxidase, and esterase-have also been re- solved into multiple molecular forms. However, each of these enzymes with broad substrate specificity is probably a family of enzymes the members of which have overlapping but distinct patterns of substrate specificity, and therefore they should not at present be classified into isozymes. The esterases, like the LDH isozymes, do indeed exhibit tissue specific patterns that arise gradually during the course of embryonic development, but the dehydrogenases owing to greater substrate speci- ficity are more suitable material for studying the origin and tissue distribution of closely related but distinct molecular species. In this report the LDH isozymes have been studied principally with special reference to their tissue and species specificity and to their ontogeny. Partly because of its use as a diagnostic tool in human disease LDH has attracted considerable attention and numerous investigators have demonstrated the oc- currence of this enzyme in multiple forms. The work of Wieland and Pfleiderer2 and of Pfleiderer and Jeckel3 has been especially noteworthy in demonstrating the species and tissue specificity of the isozymes of LDH. We have been able to con- firm their general results. However, the greater resolving power and precision of a combination of electrophoretic and histochemical techniques used in the present investigation have enabled us to amplify and clarify their results. For example, these investigators found one LDH isozyme in pig heart and two in beef heart, whereas we find three and four, respectively. Moreover, comparative electro- phoretic mobilities can be more clearly ascertained by the starch gel electrophoresis method used in the present investigation. Materials and Methods.-One of the best methods for resolving complex mixtures of proteins is zone electrophoresis in starch gels. This method developed initially by Smithies52 for separating serum proteins has proved valuable in the separation of several enzymes30' 31, 53 from tissue homogenates. The details of procedure have been set forth previously.31' 52 By these methods of starch gel electrophoresis about 1 mg of protein in a homogenate volume of 0.02 ml may be resolved in six hours. After electrophoretic resolution of the homogenate the starch block is split horizontally and one half stained with amido black to reveal the location of protein bands. The remaining half is then used for locating enzymatic activity. The identification of each kind of enzyme presents problems peculiar to itself. By the use of conventional histochemical methods several enzymes have been iden- tified directly on the starch gel-namely, esterase, phosphatases, tyrosinases, and peroxidases. The starch strip containing the visualized enzymes has been called a zymogram.30 Methods for identifying dehydrogenases directly on the starch gel all proved unsatisfactory and so an indirect method was developed that should have general applicability to any electrophoretically mobile enzyme using DPN or TPN as cofactor. In this method the starch strip, after electrophoresis, is placed in a tray and overlayed to a depth of 2 mm with an agar solution (1.5 per cent) at 450C containing the reactants. For LDH these are lactate, DPN, hydrazine, diaphorase, methylene blue, and neotetrazolium chloride. Other enzymes require somewhat different reactants. The tray is then covered quickly to exclude oxygen, cooled in an ice bath until the agar gels, and then placed in a dark incubator at 370C. Within about one hour the location of dehydrogenase activity in the starch is Downloaded by guest on September 27, 2021 VOL. 45, 1959 BIOCHEMISTRY: MARKERT AND MOLLER 755 revealed by the precipitation of a purple formazan dye in the agar directly above the area of the starch containing a dehydrogenase. A schematic representation of the reactions involved is shown in Figure 1. This method essentially detects DPNH or TPNH-hence its general applicability. The sensitivity can be very great since the accumulated product of a prolonged enzyme reaction is measured. LDH activity of less than a gamma of protein is readily detected. /D/APHORASE AGAR LACTATE HYDRAZINE NEOrETRAZOLum DPN GEL ADEJJ - STARCH (-J N(--LOH),AiLOH)U,(e- ~~~~~~~GEL ELECTROPHORETIC MIGRATION ,-DIAPHORASE MBe AtXHL DPN AGAR GEL PYRUVATE D_,PNH (TRAP) LLDHCA STARCHGE FIG. l.-Schematic representation of the reactions involved in localizing LDH isozymes after they have been separated by electrophoresis in a starch gel. A 1.5 per cent solution of agar containing the following reactants is poured over the starch and quickly cooled: sodium-DLlactate, 0.1 M; DPN, 0.3 mg/ml; hydra- zine (neutralized), O.1M; diaphorase, 0.6 units/ml. 4; neotetrazolium chloride, 0.5 mg/ml; phosphate buffer pH 7.4, 0.67 M; methylene blue, 0.003 mg/ml. This method detects DPNH or TPNH since these substances reduce neotetrazo- lium to an insoluble formazan dye. Once an enzyme has been located on one half of the starch strip the corresponding area of the other half may be cut out, frozen to break up the gel structure, and the enzyme eluted for further tests. An additional, though less satisfactory method in our hands, for locating dehydrogenases is based on the loss of fluorescence under ultraviolet light as DPNH is oxidized to DPN.2 By this method the dehydrogenases show up as dark bands on a fluorescent background. Results.-Tissue and species specific patterns of LDH isozymes: Purified, crystal- line beef heart LDH was electrophoretically resolved into five distinct protein bands, four of which had enzymatic activity. The band at position III separates into two components after prolonged electrophoresis, but is pictured as occupying a single position in Figure 2. Crude homogenates of beef heart, although containing many different proteins, were resolved into a similar pattern of LDH isozymes. Downloaded by guest on September 27, 2021 756 BIOCHEMISTRY: MARKERT AND M0LLER PROC. N. A. S. However, an isozyme from the crude homogenate appeared at position V, and this isozyme apparently was lost during the purification of the crystalline preparation. On the other hand the crystalline preparation contained a minor isozymic component appearing at position IV that was not evident in the crude homogenate. This new component may be a preparative artifact or may result from the concentration of an isozyme that is too dilute to be detected in the crude homogenate. The LDH isozyme pattern of beef heart was compared with the pattern of iso- zymes from the heart of sheep, pig, mouse, and rabbit (Fig. 2). Each pattern is unique, though the isozyme at position II is present in each species. Thus far, seven electrophoretically distinct mammalian LDH isozymes have been recognized. Two of these from beef heart are closely adjacent in position III but the remaining five occupy distinctly separated electrophoretic positions.